Xiangjie Wang

Optical flow estimation is a fundamental and long-standing task in computer vision, facilitating the understanding of motion within visual scenes. In this study, we aim to improve optical flow estimation, particularly in challenging scenarios involving small and fast-moving objects. Specifically, we proposed a learning-based model incorporating two key components: the Hierarchical Motion Field Alignment module, which ensures accurate estimation of objects of varying sizes while maintaining manageable computational complexity, and the Correlation Self-Attention module, which effectively handles large displacements, making the model suitable for scenarios with fast-moving objects. Additionally, we introduced a Multi-Scale Correlation Search layer to enhance the four-dimensional cost volume, enabling the model to address various types of motion. Experimental results demonstrate that our model achieves superior generalization performance and significantly improves the estimation of small, fast-moving objects.

To improve the accuracy of heat exchanger computational fluid dynamics (CFD) simulations, the transverse resistance coefficient of the cold side and the hot side of the water-cooled charge air cooler (WCCAC) is calculated using the fin element method, and the influence of several common factors on the resistance coefficient of the fin element is analyzed. The transverse resistance coefficient obtained from the simulation of the fin element is substituted into the WCCAC model and compared with experimental data. It was found that the fin element method can simulate the flow field of the WCCAC accurately, and the simulation results were closer to the experimental curve compared with the empirical method. This study provides guidance for the optimal design of the heat exchanger and is helpful to shorten the development time and to save costs.